List the tools that are available on Centos 6.9 Minimal.






- Goal
- Contents
- Downloadable Assets
- Project Log






available_tools_on_centos_69_minimal.txt

display_bytes.py

available_tools_on_centos_69_minimal.txt is a list of the available tools (commands, aliases, builtins, keywords) in the bash shell on Centos 6.9 Minimal. It can be searched using e.g. .

display_bytes.txt is a Python 2.7 program that reads a file (specified in the code in the variable ) and displays the contents as ASCII characters with some substitutions that facilitate reading whitespace and unprintable character bytes.






I recently succeeded in installing Centos 6.9 Minimal on Kalkin.


I would now like to know what tools are built into the bash shell on the Centos 6.9 Minimal operating system.


Press the power button to boot Kalkin.


Screen:




Screen:




Screen:




Hm.

Press the power button to turn Kalkin off.

Press the power button to boot Kalkin.

Centos boots.


Reboot five more times.

Centos boots each time.

Odd.



Press and hold the power button to turn Kalkin off.

Press the power button to boot Kalkin.


Screen:




[A booting screen is displayed]


Screen:






I'd like to record bash commands and their output into a text file. I could then mount a USB memory stick and copy this data directly into this project log, instead of transcribing it.


Google "record terminal output".


Fourth result:
askubuntu.com/questions/161935/how-do-i-log-all-input-and-output-in-a-terminal-session


Excerpt:

Q:

How do I keep logs of all the work done via terminal? For example, I want a record of all the work done on database via terminal.

edited Jul 10 '12 at 12:31
Eliah Kagan

asked Jul 10 '12 at 10:58
Ankit


A:

When you are ready to start recording a log file, type:

script screen.log

Now, until you stop the script, all input and output in the Terminal will be stored in screen.log. When you are done, just type:

exit

Your screen.log file will stored in the local directory. If you want to redirect it, use an absolute pathname such as ~/screen.log. This will do exactly what you are looking for.

Source: Ubuntu Guide - How To Log Terminal Commands [link]

edited Nov 1 '16 at 23:34
Alan Thompson

answered Jul 10 '12 at 11:21
xlukasx

There's no guarantee that something from Ubuntu will work on Centos, but in principle the idea of the script utility is simple and useful enough that it's a good bet that it will exist on many versions of Linux. It probably predates Ubuntu.





Excellent.

I'll mount a memory stick and transfer this data.


Insert memory stick (Sandisk, 8 GB) into Kalkin.




Output hangs. Press enter. New prompt appears.


Hm.

Remove memory stick.

Insert memory stick into Aineko (2008 Macbook).

Open Disk Utility.

Disk Utility reports that the memory stick's partition map scheme is Master Boot Record and its format is MS-DOS (FAT32).

Eject and remove memory stick from Aineko.


Hm.

Let's try with a different memory stick (Kingston, 32 GB).

Insert memory stick into Aineko (2008 Macbook).

Open Disk Utility.

Disk Utility reports that the Kingston memory stick's partition map scheme is Master Boot Record and its format is MS-DOS (FAT32).

Eject and remove memory stick from Aineko.


Insert this memory stick into Kalkin.



Output hangs. Press enter. New prompt appears.





Disk /dev/sdg is the expected size (~32 GB) for the memory stick.

Mount the memory stick.



The file CentOS-6.9-x86_64-minimal.iso is left over from a previous project.




Remove the memory stick.

Insert the stick into Aineko.

It mounts automatically and an icon appears on the Desktop.

Open screen.log.

It's empty.


Hm.


Eject and remove the memory stick from Aineko.

Insert the memory stick into Kalkin.



Output hangs. Press enter. New prompt appears.





Why has the contents of screen.log not been preserved?

I'll try copying to a specific filename instead of targeting a directory.





Hm. It looks as though the data has been successfully copied onto the disk.





Remove the memory stick from Kalkin.

Insert the memory stick into Aineko.

It mounts automatically and an icon appears on the Desktop.


Open foo.txt.

Contents (as shown in TextWrangler):



I don't why there are extra newlines.


Let's look at the raw bytes in foo.txt.


I'll write some Python code to do this.





Move foo.txt to Downloads directory.

Save the Python code in the Downloads directory as display_bytes.py.





There are [ord_13] bytes in all the locations where I saw an extra newline in TextWrangler's rendering of foo.txt.

Looking at my ASCII reference table, I find that byte 13 is CR (carriage return). Note: The carriage return is often typed in e.g. Python strings as "\r".

Evidently TextWrangler displays carriage returns while ignores them.

I'm slightly surprised that outputs carriage return bytes.


Delete foo.txt and screen.log from the memory stick.

Delete foo.txt from the Downloads directory on Aineko.

Eject and remove the memory stick from Aineko.


I'll stop here for today.

Shut down Kalkin.




Another day, another crack at the problem.


Press the power button to boot Kalkin.


Screen:




Screen:




Screen:




Hm.

Press Enter.

Screen:




Press the power button to turn Kalkin off.


Press the power button to boot Kalkin.

Centos boots.

Press and hold the power button to turn Kalkin off.


Reboot four more times.

Centos boots each time.


It's as if the first boot changes a setting somewhere, which is stored in a capacitor, the charge in which then fades when the computer is off for a reasonably long period of time.


Press and hold the power button to turn Kalkin off.

Press the power button to boot Kalkin.


Screen:




[A booting screen is displayed]


I want to look at this problem again:
- Copying a file to a target directory (instead of a target filename) on a mounted memory stick results in an empty 0-byte file.


Screen:





Insert memory stick (Kingston, 32 GB) into Kalkin.




Output hangs. Press enter. New prompt appears.


I'd like to find the device identifier of the inserted memory stick via reading something more concise than the output of .


While installing Centos 6.9 Minimal on Kalkin, I discovered the command.



The Kingston stick has the identifier .




Hm. Ah, maybe the trailing slash on the target directory makes a difference. I used one before when the command produced a 0-byte file.





Hm.




Remove the memory stick.

Insert it into Aineko.

It mounts automatically and an icon appears on the Desktop.





Hm. Neither file is present.

Eject and remove the memory stick from Aineko.


Insert the memory stick into Kalkin.



Output hangs. Press enter. New prompt appears.


Running shows me that the identifier is again .





Neither file is present.

Hm. Let's try again, but this time I'll dismount it and remount it without removing it from Kalkin.

Hm. [some reading occurs here]. Wait a second, I've been using incorrectly. One should target the mountpoint directory, not the device identifier.




It looks like the problem was caused by incorrect use of .




Remove the memory stick from Kalkin.

Insert the memory stick into Aineko.

It mounts automatically and an icon appears on the Desktop.





Excellent.


Eject and remove memory stick from Aineko.

Insert memory stick into Kalkin.




Output hangs. Press enter. New prompt appears.





When I was checking Kalkin's hardware details using a Centos 6.9 bootable USB stick, I discovered the command, a bash builtin, which generates possible completion matches for words in the bash shell. If no words are provided, and it is given at least one option, it will list all possible words (given that option).


I'm going to run with various options in order to gather lists of the available commands on Centos 6.9 Minimal.


I will use the memory stick to transfer the results, recorded via , to Aineko for inclusion in this log.





[a shutdown occurs here. I've rerun the appropriate commands and arrived back at this point.]





[various commands are run here. output will be pasted into this log later.]





Remove the memory stick from Kalkin.

Insert the stick into Aineko.

It mounts automatically and an icon appears on the Desktop.


The file work.log is present but it's 0 KB i.e. empty.

Hm.

Eject and remove memory stick from Aineko.

Insert the stick into Kalkin.




Output hangs. Press enter. New prompt appears.


I'll copy to a filename rather than to a directory.




Remove the memory stick from Kalkin.

Insert the stick into Aineko.

It mounts automatically and an icon appears on the Desktop.


work.log is again 0 KB.


Looking back at this log, it appears that when I followed this sequence:
- unmount memory stick on Kalkin
- mount memory stick on Kalkin
- unmount memory stick on Kalkin
and only then removed the memory stick and inserted it into Aineko, the data was preserved. Odd.

Let's try that again.


Eject and remove memory stick from Aineko.

Insert the stick into Kalkin.




Output hangs. Press enter. New prompt appears.





Remove the memory stick from Kalkin.

Insert the stick into Aineko.

It mounts automatically and an icon appears on the Desktop.


work.log is present and its size is 131 KB.

The unmount and remount step seems to do something that causes the data to be preserved on the memory stick.






On opening the file in TextWrangler, I see the extra newlines caused by the inclusion of carriage return bytes.


I'd like to remove the carriage return bytes from the output.


Google "count occurrences of particular byte".

Second result:
stackoverflow.com/questions/1603566/count-occurrences-of-a-char-in-plain-text-file


Excerpt:

Q:

Is there any way under linux/terminal to count, how many times the char f occurs in a plain text file?

edited Oct 21 '09 at 21:24
Rob Hruska

asked Oct 21 '09 at 20:56
cupakob


A:

[...]

tr -cd f < file | wc -c

[...]

edited Jan 29 '13 at 10:05
erik

answered Jan 17 '13 at 0:33
user1985553


Comment:

To count several characters, e.g.

a

,

b

and

c

:

tr -cd abc < file | wc -l

.

Skippy le Grand Gourou Apr 3 '17 at 13:26

Well, I know from running that work.log is 9072 lines long and therefore contains 9072 newline bytes. This is a data point.



Cool. Evidently there's some rules about quoting and backslash characters.




4032 + 1431 = 5463

I'm not sure why this addition didn't hold for \, n, and \n, (not quoted, so treated as separate bytes?) in the earlier commands. Probably backslash is special.





9069 = 9072 - 3

This carriage return byte count matches the first script recording, in which the first line and the last two lines each had only a newline, instead of carriage return + newline.





Excerpt from man page:

NAME

tr -- translate characters

SYNOPSIS

tr [-Ccsu] string1 string2
tr [-Ccu] -d string1
tr [-Ccu] -s string1
tr [-Ccu] -ds string1 string2

DESCRIPTION

The tr utility copies the standard input to the standard output with substitution or deletion of selected characters.

The following options are available:

-C
Complement the set of characters in string1, that is ``-C ab'' includes every character except for `a' and `b'.

-c
Same as -C but complement the set of values in string1.

-d
Delete characters in string1 from the input.

[...]

In the second synopsis form, the characters in string1 are deleted from the input.

[...]

The following conventions can be used in string1 and string2 to specify sets of characters:

character
Any character not described by one of the following conventions represents itself.

\character
A backslash followed by certain special characters maps to
special values.
\a <alert character>
\b <backspace>
\f <form-feed>
\n <newline>
\r <carriage return>
\t <tab>
\v <vertical tab>
A backslash followed by any other character maps to that character.

[...]

EXAMPLES

The following examples are shown as given to the shell:

Create a list of the words in file1, one per line, where a word is taken to be a maximal string of letters.

tr -cs "[:alpha:]" "\n" < file1

[...]

Strip out non-printable characters from file1.
tr -cd "[:print:]" < file1

[...]

BSD July 23, 2004

Hm. From the man page descriptions, -C and -c appear to be the same option. I suspect that the description of -C should actually be "include only characters in the set of values in string1".

In the examples, I note the use of double quotation marks around string1 and string2. Evidently they can be omitted if the input string consists of non-backspace-escaped characters.

No mention of how to target the backslash character when using .

Hm. "A backslash followed by any other character maps to that character." Perhaps in the command , the target string was actually "\ ", i.e. the backslash and also the space following it, which would then be treated as a space.




Yup. It counted spaces.

Further thoughts on "A backslash followed by any other character maps to that character":
- targeting \n without quotation marks actually counted the occurrences of n by itself, which is why the results for \n and n were the same.
- to target a backslash, I should probably use "\\".


In the command:

proceeds through each line in the input (which in this case is the file work.log), copying the result of evaluating that line to the standard output.
selects every byte that is not "\r".
deletes every selected byte.

Adding to this command uses the (which stands for "word count") utility to count each character that sent to the standard output.

So, in this result:



I can see that the carriage return byte "\r" occurs 9069 times in work.log.


If I can correctly count a byte type, this means I can select it, which means I can delete it.



There are 124870 total in work.log.


I'll leave out the option, so instead should delete every instance of "\r" and copy all other bytes to the standard output.



124870 - 9069 = 115801

115801 is the number of bytes expected after removing all instances of "\r".


Finally, I'll redirect the output into a file. This file will be the same as work.log, except that it will not contain any carriage return bytes.




Excellent.


Here are the relevant excerpts from work2.log, reordered and edited a bit.


There are a few strange renderings of the command inputs e.g. and . I'm not sure why these have occurred. They are in work.log as well as work2.log. I'll remove them manually.

I've rerun the commands on Kalkin, piping the output into , and checked that the output matches that shown here. I didn't find that any of the output was rendered differently, but it's good to be sure.


Descriptions of the commands:

will list all the commands you could run.
will list all the aliases you could run.
will list all the builtins you could run.
will list all the keywords you could run.
will list all the functions you could run.
will list all the above in one go.





After listing the available commands with , I then did some exploration of the filesystem, using e.g. , etc.

Unfortunately the output produced by was rather strange, so I haven't included any here, except for the excerpt below that demonstrates the issue:

Excerpt:





Last thing: Create a searchable list of the available tools (commands, aliases, builtins, keywords) in the bash shell on Centos 6.9 Minimal.

Manually copy the outputs of:
-
-
-
-
into a new file. Remove input lines and empty lines. Save it as tmp.txt in Downloads directory on Aineko.

I'll sort tmp.txt and remove duplicate entries.




I'll add available_tools_on_centos_69_minimal.txt and display_bytes.py as downloadable assets at the beginning of this article.


Tidy up:
- On the Kingston memory stick, delete work.log and work2.log.









Changes from the original text:

- I have not always preserved the format of any excerpts from webpages on other sites (e.g. not preserving the original bold/italic styles, changing the list structures, not preserving hyperlinks).

- I have not always preserved the format of any computer output (e.g. from running bash commands). Examples: Setting input lines in bold text, adding/removing newlines in order to make a sequence of commands easier to read, using hyphens for lists and sublists instead of indentation, breaking wide tables into consecutive sections.